In intraoperative image guided surgery, the surgeon's view is enhanced by augmenting standard camera views with advanced medical images (e.g., PET, CT, flouroscopy, and MRI). The medical images assist in the execution and evaluation of the surgery. Of the various medical imaging modalities, Magnetic Resonance Imaging (MRI) is the optimal choice for high quality soft tissue imaging. However, the strong magnetic fields and constrained workspaces within most MRI machines limits the use of MRI as an interoperative imaging tool. Recent innovations in MRI compatible surgical robots have enabled MRI-guided surgeries that use actuated tools inside the constrained MRI environment. The large size of these robots, however, limits the accessible regions of the body during MRI scans. We propose to create a small size, high performance system for actuating tools inside MRI machines with no effect on the image quality. Such tools will increase the number of surgeries that can be enhanced through intraoperative MRI scans, improving surgical safety and reducing surgical duration. We will use mechanical cable drive technology to attain the high performance of DC motors, while maintaining a small sized, MRI compatible actuator. Our mechanism will be the first to allow precision actuated tool use at any location in the MRI bore, without requiring an open field MRI. The aims of this project are: 1. Develop a completely non-metallic MRI compatible actuation mechanism. A high performance, small footprint, non-metallic MRI compatible actuator will allow surgeons to derive all of the image quality of MRI with intraoperative image guided surgery. 2. Develop control algorithms for cable drive actuators. The main technological hurdles to using cable driven actuators in a precision environment such as surgery are: cable stretch, which complicates the relationship between motor torque and actuated axle motion, and cable friction, which results in transmission losses and hysteresis effects. We will use advanced control techniques to overcome these challenges. [unreadable] [unreadable] [unreadable]